[PATCH 3/6] 8xx: invalidate non present TLBs

Benjamin Herrenschmidt benh at kernel.crashing.org
Fri Oct 9 07:44:13 EST 2009

> One could try clearing the store bit in the page fault handler, but then
> that might cause a loop.
> Not sure it has any practical meaning though.
> Anyhow, you are welcome to have a look at the patches I have been tossing out.

The store bit in do_page_fault() is -very- important (and the only DSISR
bit that is as I wrote earlier). The generic code must know if a fault
is caused by a load or a store since the later is what will cause
copy_on_write to happen for example, or dirty to be set, etc..


More information about the Linuxppc-dev mailing list